Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2024 Apr 1.
Published in final edited form as: Psychoneuroendocrinology. 2023 Jan 5;150:106024. doi: 10.1016/j.psyneuen.2023.106024

Employment and family income in psychological and immune outcomes during bereavement

Jensine Paoletti a,*, Michelle A Chen b, E Lydia Wu-Chung a, Ryan L Brown c, Angie S LeRoy d, Kyle W Murdock e, Cobi J Heijnen f, Christopher P Fagundes a,f,g
PMCID: PMC9974808  NIHMSID: NIHMS1867510  PMID: 36702040

Abstract

Spousal bereavement is one of the most stressful experiences in adulthood. In a sample of 183 widow(er)s, bereaved about three months prior, we examined the intersection of employment, family income, and health outcomes (proinflammatory marker production, perceived stress, and grief symptoms). Bereaved employees had higher levels of monocyte-stimulated interleukin-6, tumor necrosis factor-α, chemokine ligands 4, and perceived stress than bereaved retirees. We also found an interaction such that family income was positively associated with perceived stress and grief symptoms for employed window(er)s, but not for retirees. These findings align with the reserve capacity model, which states that people at higher levels of socioeconomic status have more psychosocial resources to address psychosocial stressors. Employment likely served as an added psychological and inflammatory burden for all bereaved workers, except those with the highest incomes.

Keywords: bereavement, work-family, retirement, psychosocial stress

1. Introduction

The death of a spouse is a major life stressor requiring significant readjustment, especially in the first year of widowhood (Holmes & Rahe, 1967). According to the dual process model of loss, widow(er)s must grieve the loss, while simultaneously confronting several novel stressors (e.g., self-esteem, identity, daily routine, and social networks) that directly or indirectly result from the spouse's absence (Stroebe et al., 2007). The stressors that characterize the aftermath of a spouse's death may partly contribute to an excess risk of morbidity and premature mortality in early widowhood (Stahl et al., 2016). Psychological stress can promote chronic low-grade inflammation (Rohleder, 2014). While acute inflammation in response to a pathogen or injury is an adaptive and necessary immune response, chronic inflammation is associated with chronic illness and is a significant risk factor for premature mortality (Fagundes & Wu, 2020).

Although widow(er)s exhibit higher levels of chronic low-grade inflammation than matched comparisons (Fagundes et al., 2018), there is considerable heterogeneity. Indeed, some widow(er)s are remarkably resilient, while others have health difficulties (Fagundes & Wu, 2020). Understanding risk patterns is critical to designing targeted interventions and may also have policy implications. Researchers have identified pre-loss factors, such as early life trauma and psychiatric history of depression, as predictors of poor mental health outcomes (Fagundes & Way, 2014). Unfortunately, we know less about how post-loss context impacts risk. In addition, most of the work predicting risk and resilience has focused on mental health, rather than looking “under the skin” at objective markers of physical health. Identifying post-loss contextual risk factors for physical health problems is critical to designing effective, personalized interventions.

Work-related hassles are a major source of daily stress for most adults (Goh et al., 2016). When widow(er)s are employed, they must grieve their spousal loss and cope with typical workplace stressors (Fitzpatrick & Bossé, 2001). Because of the additional stressors associated with employment (e.g., low autonomy, high demands, low social support at work), employed widow(er)s may be at particular risk for mental and physical health problems. Alternatively, employment can have positive health benefits in the form of social support, positive challenges, and sometimes status (Erdogan et al., 2012). Widow(er)s who derive a comfortable lifestyle from their job may not be as negatively impacted by the additional stress of employment. Indeed, having a higher income may afford them conveniences that offset the stress associated with the additional employment-related stress. However, for those who are working out of financial necessity or even to maintain a middle-class lifestyle, the stress associated with being employed after the death of a spouse may have mental and physical health repercussions.

Income drives disparate health outcomes after bereavement (Bindley et al., 2019). Financial concerns are paramount in studies of widowhood, particularly among spouses who provided unpaid care to their late spouse (DiGiacomo et al., 2015; Roulston et al., 2017; Stahl et al., 2016). In one bereavement study, income directly predicted psychological health among bereaved spouses (Utz et al., 2011). Thus, we expect a family income to interact with employment status for proinflammatory cytokine production, perceived stress, and grief symptoms. Widowed persons with higher family incomes be less dependent on any one source of income (i.e., the surviving spouse's job). Therefore, the surviving spouse is less likely to feel they must work to support themselves. The additional family income may give them a sense of security and access to more resources, including taking unpaid bereavement leave. Finally, differences in income are likely more meaningful for employed versus retired people, as retired people often live on a fixed income and must have sufficient financial resources to retire in the first place.

We expect that employed widow(er)s will experience worse health outcomes than their retired counterparts as they must cope with the loss of their spouse and ensuing demands (i.e., settling affairs) while managing the demands of their job. Specifically, we hypothesize proinflammatory cytokine production, perceived stress, and grief symptoms to be higher among employed widow(er)s compared to retired widow(er)s due to demands in two domains, work and family. We will measure interleukin-6 (IL-6), tumor necrosis factor-α (TNFα), and chemokine ligands 4 (CCL4). We hypothesize employed widow(er)s with lower and middle levels of family income will have higher levels of (a) proinflammatory marker production, (b) perceived stress, and (c) grief symptoms than employed widow(er)s with higher family incomes and retired widow(er)s. Through testing these hypotheses, we contribute to the literature in three main ways. First, we examine the effects of a pre-loss factor, employment status, in a bereavement context. Next, we examine the intersection of employment status and socioeconomic status. Finally, we add to the body of research using an occupational health psychology perspective to examine widowhood.

2. Materials and Methods

2.1. Participants and Procedure

Data in this study represents secondary analysis from a project designed to examine low-grade inflammation among widow(er)s during the first year after their bereavement. We examined the relationship between employment status (employed full-time or part-time versus retired) and several psychological and inflammation-related variables in a sample of 183 bereaved spouses (M = 68.57 years; 67% women; see Table 1 for all sample characteristics). Inclusion criteria for this study necessitated participants were bereaved in the last three months (M = 82.83 days), were married to their late spouse for at least three years, and spoke English. Participants were excluded if they experienced an additional loss of a loved one in the last year, were divorced in the last year, were undergoing cancer treatment, had a pacemaker, or had an autoimmune disease. Participants completed data collection in the lab or at a home visit at about three months post-bereavement. Data collection involved demographic and self-report questionnaires as well as blood draws to assess proinflammatory marker production. To minimize confounding effects on inflammation, participant visits were scheduled in the morning, and participants were instructed to avoid strenuous exercise 48 hours before their visit and reschedule visits if they were feeling symptoms of an acute illness. Finally, participants were provided a list of high fat foods that they should avoid on the morning of their visit (e.g., butter), in addition to avoiding caffeinated beverages.

Table 1.

Sample characteristics.

Variable Count (%) or M (SD)
Overall Employed (n = 64) Retired (n = 119) p-value
Age (years) 68.57 (8.84) 62.11 (9.34) 72.03 (6.35) < .001***
Racea – – – .85
 White 165 (90.16%) 58 (90.63%) 107 (89.9%) –
 Black 13 (7.10%) 5 (7.81%) 8 (6.72%) –
 Asian 3 (1.64%) 1 (1.56%) 2 (1.68%) –
 Other 2 (1.09%) 0 (0%) 2 (1.68%) –
Sex (female)b 122 (66.67%) 39 (60.94%) 83 (69.75%) .25
Subjective social status 6.63 (1.79) 6.86 (1.73) 6.51 (1.81) .21
Educationa – – – .015*
 Less than high school 1 (0.5%) 0 (0%) 1 (0.84%) –
 High school diploma 37 (20.22%) 10 (15.63%) 27 (22.69%) –
 Associate’s degree 21 (11.48%) 4 (6.25%) 17 (14.29%) –
 Bachelor’ s degree 62 (33.88%) 22 (34.38%) 40 (33.61%) –
 Master’s degree 40 (21.86%) 17 (26.56%) 23 (19.33%) –
 Doctorate or professional degree 22 (12.02%) 11 (17.19%) 11 (9.24%) –
Bereavement length (days) 82.83 (17.16) 83.71 (17.70) 82.19 (16.88) .50
Duration of marriage (years) 36.52 (16.03) 27.79 (15.28) 40.59 (15.28) < .001***
Marital satisfaction 4.11 (0.96) 4.11 (0.95) 4.12 (0.98) .96
Body mass index 28.34 (5.64) 27.57 (5.67) 28.84 (5.62) .14
Alcoholic drinks per week 3.39 (5.19) 2.90 (3.31) 3.65 (5.94) .35
Sleep quality 8.16 (4.16) 7.90 (4.13) 8.30 (4.19) .55
Comorbidities 0.40 (1.29) 0.17 (0.46) 0.52 (1.55) .08
Anti-inflammatory medications 0.69 (0.46) 0.52 (0.50) 0.78 (0.41) < .001***
Family incomea – – – < .001***
 Less than $ 5000 1 (0.60%) 0 (0%) 1 (0.97%) –
 $12,000 - $15,999 1 (0.60%) 0 (0%) 1 (0.97%) –
 $16,000 - $24,999 4 (2.40%) 0 (0%) 4 (3.88%) –
 $25,000 - $34,999 7 (4.19%) 2 (3.13%) 5 (4.85%) –
 $35,000 - $49,999 16 (9.58%) 3 (4.69%) 13 (12.62%) –
 $50,000 - $74,999 25 (14.97%) 7 (10.94%) 18 (17.48%) –
 $75,000 - $99,999 29 (17.37%) 9 (14.06%) 20 (19.42%) –
 $100,000 + 84 (50.30%) 43 (67.19%) 41 (39.81%) –
Grief symptoms 10.28 (12.33) 11.14 (11.96) 9.77 (12.62) .43
Perceived stress 0.12 (0.73) 0.28 (0.78) 0.01 (0.68) .01*
Composite inflammation −0.21 (1.04) −0.07 (0.91) −0.29 (1.09) .19
 IL-6 −0.22 (1.18) −0.08 (1.08) −0.29 (1.23) .25
 TNF-α −0.22 (1.04) −0.07 (0.95) −0.30 (1.08) .16
 CCL4 −0.21 (1.05) −0.08 (0.88) −0.27 (1.12) .23
Open in a new tab

p-values derived from one-way ANOVA except where otherwise specified.

a

indicates p-value derived from Wilcoxon-Mann-Whitney test.

b

indicates p-value derived from chi-square test of independence.

*

p < .05.

***

p < .001.

2.2. Measures

2.2.1. Monocyte Stimulated Proinflammatory Markers

Activation of the stress response system can be measured with the proinflammatory markers produced by monocytes (i.e., macrophages), a type of white blood cell. We derived monocyte-stimulated interleukin-6 (IL-6), tumor necrosis factor-α (TNFα), and chemokine ligands 4 (CCL4) from whole blood. Interleukin-6 (IL-6) and tumor necrosis factor-α (TNFα) are reliably responsive to psychological stress (Bruenig et al., 2014; Fagundes et al., 2018; Gouin et al., 2009; LeRoy et al., 2020; Pawlowski et al., 2014; Schultze-Florey et al., 2012). Chemokine ligands 4 (CCL-4) had high internal consistency with IL-6 and TNFα, making a reliable index (α = .91). Proinflammatory marker production was induced in whole blood cell cultures to measure the reactivity of leukocytes to lipopolysaccharide (LPS). We used heparinized whole blood diluted 1:10 with RPMI-1640 (Gibco) to induce LPS-stimulated cytokine production and stimulated with 1 ng/mL LPS (Sigma) at 37 °C and 5% CO2 for 24 h. After 24 h of culture, we collected supernatants and stored at −80 °C until we analyzed them using multiplex assays per the manufacturer's instructions (R&D Biosystems). All procedures were based on previous work (Fagundes et al., 2018; Zuiden et al., 2011).

2.2.2. Perceived Stress

Perceived stress was measured with the 10-item Perceived Stress Scale, which asks participants to describe how often they have felt stressed in the last month (Cohen et al., 1983). Participants responded on a scale of 0 (never) to 4 (very often). An example item reads, "How often have you been upset because of something that happened unexpectedly?" The items were reverse-coded when appropriate, mean scored, and centered (α = .81).

2.2.3. Grief Symptoms

Grief symptoms were measured with the 15-item Utrecht Grief Rumination Scale (Eisma et al., 2014), which measures ruminative coping, depressive rumination, and grief rumination. Grief rumination is thought to lead to prolonged activation of the hypothalamic–pituitary–adrenal (HPA) axis over time (O’Connor, 2019; Zoccola & Dickerson, 2012). Participants responded on a scale of 0 (never) to 4 (very often), reflecting on their experiences in the last month. For example, one item asks how often participants "think about the consequences that [the spouse's] death has for you." The items were mean scored and centered (α = .89).

2.2.4. Employment

Employment status was assessed with a self-report question. Participants selected the category that best fit the following options: full-time for pay, part-time for pay, retired, disabled, unemployed, or other. As we were interested in broader questions of employment versus retirement (n =119), we collapsed the full-time for pay (n = 44) and part-time for pay (n = 20) groups together to create the employed group. We excluded the additional groups, including disabled (n = 1), unemployed (n = 7), and other (n = 10) as they were outside of our research question. We dummy-coded the employment status variable such that 0 represented retired participants and 1 represented employed participants.

2.2.5. Family Income

Family income was measured with an item from the Sociodemographic Questionnaire (Adler et al., 2000). The item asked participants to categorize their family's total, pre-tax income from the last 12 months, including wages, rent from properties, social security, disability and/or veteran's benefits, unemployment benefits, workman's compensation, help from relatives (including child payments and alimony), and other sources. Responses were recorded as one of nine ordinal categories, from 0 (less than $5,000) to 8 ($100,000 or more). We created an interaction term by multiplying employment status and family income.

2.2.6. Covariates

According to field recommendations, demographic and inflammation-related health information were included as covariates in analyses (O'Connor et al., 2009). Participants self-reported age, race, sex, education, number of weekly alcoholic drinks, and length of bereavement in days. Participants reported education (highest degree earned) and their subjective social status within their community (on a scale of 1 to 10) as part of the Sociodemographic Questionnaire (Adler et al., 2000). Participants' height and weight were gathered through in-person measurements to calculate body mass index (BMI). Sleep quality was measured with the Pittsburgh Sleep Quality Index total score, containing 20 items (Buysse et al., 1989). Comorbidities were measured with the Charlson Comorbidity Index, which assigns weights to 19 comorbid conditions based on their potential influence on one-year mortality (Charlson et al., 1987). We included a count of the inflammation-related medications (i.e., aspirin, statins, antihypertensive, and antidepressants; Kenis & Maes, 2002; Lin et al., 2020).

2.3. Analyses

We log-transformed the inflammatory markers, IL-6, TNF-α, and CCL4, to account for the expected skewness. We z-score transformed the data and created a composite index of the proinflammatory markers (α = .91) to reduce the possibility of a Type I error, aligned with past research (e.g., LeRoy et al., 2020). We conducted three linear regressions in SPSS, accounting for relevant demographic and health-related covariates (O'Connor et al., 2009). In all models, we removed missing data listwise. In the first hierarchical regression model, we predicted the proinflammatory composite variable with employment status and additional covariates, including age, BMI, biological sex, weekly alcoholic drinks, sleep quality, inflammation-related medications, comorbidities, education, and subjective social status. Next, we ran a hierarchical linear regression predicting perceived stress and an additional regression model predicting grief symptoms. For both psychological variables (i.e., perceived stress and grief), the regressions included predictors of employment status, family income, the interaction variable between family income and employment status, as well as the following a priori covariates: age, sex, and sleep quality. Finally, we included exploratory post-hoc analyses exploring the association of duration of marriage (self-reported in years), marital satisfaction (measured with one item from the Couples Satisfaction Index; Funk & Rogge, 2007), and biological sex on proinflammatory markers, perceived stress, grief symptoms. It is worth noting that retrospective reports of marital satisfaction after bereavement are inflated relative to marital satisfaction prior to bereavement (Mancini et al., 2009). For exploratory post-hoc analyses, we examined the direct associations, all possible 2-way interactions with employment status and family income, and the 3-way interaction of employment status, family income, and duration of marriage.

3. Results

3.1. Proinflammatory Marker Production Results

Sample descriptive statistics can be found in Table 1. To test Hypothesis 1a (the effect of employment on proinflammatory marker production) and Hypothesis 2a (the interaction of employment and family income on proinflammatory marker production), we conducted a hierarchical linear regression predicting the proinflammatory composite. This regression included relevant covariates from the literature known to affect inflammatory markers, specifically age, BMI, sex, alcoholic drinks per week, sleep quality, inflammation-related medications (i.e., a dummy variable to indicate if participants were taking aspirin, statins, antihypertensives, or antidepressants), comorbidities (i.e., a count of the number of comorbid conditions, such as diabetes; O'Connor et al., 2009), education, and subjective social status. Our results supported Hypothesis 1a, as employment status was predictive of the proinflammatory composite such that employed participants had higher proinflammatory marker production than retired participants (b = 0.47, SE = 0.21, p = .029; see Table 2 and Figure 1). Follow-up analyses indicate that employment status was significantly related to all three included proinflammatory markers, IL-6 (b = 0.51, SE = 0.24, p = .039), TNF-α (b = 0.47, SE = 0.21, p = .03), and CCL4 (b = 0.43, SE = 0.22, p = .046). Our results did not support Hypothesis 2a, as the interaction between employment status and family income was not significantly predictive of proinflammatory marker production (b = −0.16, SE = 0.14, p = .27); therefore, Table 2 only reflects models relevant to the main effects.

Table 2.

Hierarchical regression for composite proinflammatory marker production.

B 95% CI for B SE B β R 2 ΔR2
LL UL
Model 1: Covariates .02 .024
(Constant) −1.30 −3.31 0.72 1.02 –
Age 0.01 −0.01 0.03 0.01 0.08
BMI 0.01 −0.03 0.04 0.02 0.03
Sex2 0.00 −0.37 0.37 0.19 0.00
Alcoholic Drinks Per Week −0.001 −0.03 0.03 0.02 −0.01
Poor Sleep Quality −0.01 −0.05 0.04 0.02 −0.02
Inflammatory Medications 0.20 −0.21 0.60 0.20 0.09
Comorbidities −0.05 −0.17 0.08 0.07 −0.06
Education 0.23 −0.11 0.17 0.07 0.04
Subjective Social Status 0.01 −0.08 0.11 0.05 0.02
Model 2: Main Effects .07 .044*
(Constant) −3.03* −5.40 −0.66 1.20
Age 0.03* 0.00 0.05 0.01 0.22
BMI 0.01 −0.02 0.04 0.02 0.04
Sex2 0.07 −0.29 0.43 0.18 0.03
Alcoholic Drinks Per Week 0.01 −0.04 0.04 0.02 0.05
Poor Sleep Quality 0.001 −0.04 0.04 0.02 0.01
Inflammatory Medications 0.23 −0.16 0.64 0.20 0.10
Comorbidities −0.04 −0.16 0.09 0.06 −0.05
Education −0.02 −0.17 0.12 0.07 −0.03
Subjective Social Status −0.01 −0.10 0.10 0.05 −0.01
Employment Status3 0.48* 0.05 0.90 0.22 0.22
Family Income 0.07 −0.05 0.19 0.06 0.10
Open in a new tab
1

CI = confidence intervall; LL = lower limit; UL = upper limit.

2

indicates dummy coding so that 0 = male and 1 = female.

3

indicates dummy coding so that 0 = retired and 1 = employed.

*

indicates p < .05.

**

indicates p < .01.

***

indicates p < .001.

Figure 1.

Figure 1

Open in a new tab

Effect of employment on proinflammatory marker production. Error bars represent standard error.

3.2. Perceived Stress Results

To test Hypothesis 1b (the effect of employment on stress) and Hypothesis 2b (the interaction of employment and family income on stress), we conducted a hierarchical linear regression predicting perceived stress. The regression included predictors of employment status, family income, the interaction between employment status and family income, as well as covariates of age, sex, and sleep quality. Our results supported Hypothesis 1b, as there was a significant association of employment status and perceived stress, such that employed people had higher perceived stress (b = 0.29, SE = 0.12, p = .012; see Table 3). Our results also supported Hypothesis 2b, as there was a significant interaction such that employed people with lower family incomes had the highest levels of perceived stress (b = −0.17, SE = 0.08, p = .029; see Figure 2). In simple slope analyses, we found that among employed people, income was negatively associated with perceived stress (b = −0.18, SE = 0.07, p = .012), while income was not associated with perceived stressed among retired people (b = −0.01, SE = 0.03, p = .89).

Table 3.

Hierarchical regression for grief symptoms and perceived stress.

Grief Symptoms Perceived Stress
B 95% CI for B SE B β R2 ΔR2 B 95% CI for B SE B β R2 ΔR2
LL UL LL UL
Model 1: Covariates .20 .20*** .38 .38***
(Constant) 20.01 5.26 34.77 7.47 – 0.66 −0.13 1.45 0.40 –
Age −0.25 −0.45 −0.06 0.10 −0.18 −0.02*** −0.03 −0.01 0.01 −0.25
Sex2 −2.55 −6.17 1.07 1.83 −0.10 0.19 −0.00 0.39 0.10 0.13
Poor Sleep Quality 1.13 0.71 1.56 0.21 0.38 1.11*** 0.68 1.53 0.21 0.38
Model 2: Main Effects .20 .00 .40 .02*
(Constant) 24.45 5.60 43.29 9.54 – 0.06 −0.93 1.05 0.50 –
Age −0.31 −0.54 −0.07 0.12 −0.22 −0.01* −0.03 −0.00 0.01 −0.16
Sex2 −2.72 −6.39 0.95 1.86 −0.11 0.22* 0.03 0.42 0.10 0.15
Poor Sleep Quality 1.11 0.68 1.54 0.22 0.38 0.09*** 0.07 0.11 0.01 0.50
Family Income −1.01 −5.32 3.30 2.18 −0.04 −0.04 −0.10 0.02 0.03 −0.08
Employment Status3 −0.50 −1.70 0.69 0.60 −0.06 0.27* 0.04 0.50 0.12 0.18
Model 3: Interaction .22 .02 .42 .02*
(Constant) 25.84 7.10 44.58 9.49 – 0.14 −0.84 1.12 0.50 –
Age −0.32 −0.56 −0.08 0.12 −0.23 −0.01* −0.03 −0.00 0.01 −0.17
Sex2 −3.21 −6.88 0.47 1.86 −0.13 0.19* 0.00 0.39 0.10 0.13
Poor Sleep Quality 1.08 0.66 1.51 0.22 0.37 0.09*** 0.07 0.11 0.01 0.49
Family Income 19.67 −1.72 41.07 10.83 0.78 −0.01 −0.07 0.07 0.04 −0.01
Employment Status3 0.09 −1.24 1.41 0.67 0.01 1.48* 0.36 2.60 0.57 0.98
Employment Status3 × Family Income −2.90 −5.84 0.04 1.49 −0.86 −0.17* −0.32 −0.02 0.09 −0.84
Open in a new tab
1

CI = confidence intervall; LL = lower limit; UL = upper limit.

2

indicates dummy coding so that 0 = male and 1 = female.

3

indicates dummy coding so that 0 = retired and 1 = employed.

*

indicates p < .05.

**

indicates p < .01.

***

indicates p < .001.

Figure 2.

Figure 2

Open in a new tab

Interaction between family income and employment status on perceived stress. Error bars represent the standard error.

3.3. Grief Symptoms Results

To test Hypothesis 1c (the effect of employment on grief) and Hypothesis 2c (the interaction of employment and family income on grief), we conducted a hierarchical linear regression predicting grief symptoms. The regression included predictors of employment status, family income, the interaction between employment status and family income, as well as covariates of age, sex, and sleep quality. Our results did not support Hypothesis 1c, as there was no significant association between employment status and grief symptoms (b = −0.70, SE = 2.19, p = .75; see Table 3). The two-tailed test results for Hypothesis 2c lean towards significance, such that employed widow(er)s with lower family incomes had the highest levels of grief symptoms (b = −2.91, SE = 1.49, p = .052; see Figure 3). Income was negatively associated with grief symptoms among employed people (b = −2.83, SE = 1.32, p = .034); however, income was not associated with grief symptoms among retired people (b = 0.08, SE = 0.67 p = .90).

Figure 3.

Figure 3

Open in a new tab

Interaction between family income and employment status on grief symptoms. Error bars represent the standard error of the mean.

3.4. Post-Hoc Results

We examined the association of duration of marriage, marriage satisfaction, and biological sex as a direct association and as a moderator. We analyzed each variable as a moderator in a 2-way interaction with employment status, in a 2-way interaction with family income, and in a 3-way interaction of employment status and family income. Caution is warranted when interpreting the results, as the number of tests may elevate the Type 1 error rate. Significant post-hoc results may be found in the Supplementary Materials.

There were no significant direct associations or 3-way or 2-way interactions of duration of marriage on proinflammatory marker production, perceived stress, or grief symptoms. There were no significant association of marriage satisfaction on proinflammatory marker production or perceived stress. There was an association of marital satisfaction on grief symptoms (b = 2.93, SE = 0.90, p = .001); the significance levels of all other effects remained the same. There were no significant 3-way or 2-way moderation of marriage satisfaction on perceived stress or grief symptoms. There was a significant 2-way interaction of marital satisfaction and family income on proinflammatory marker production (b = 0.13, SE = 0.06, p = .031), however, the simple slopes are not significant at one standard deviation above mean levels of income (b = −0.10, p = .41) or at one standard deviation below mean levels of income (b = 0.17, p = .21). There were no significant associations or 3-way or 2-way interactions of biological sex on proinflammatory marker production, perceived stress, or grief symptoms.

4. Discussion

In the present study, we found group differences in health based on employment status, such that employed widow(er)s had higher levels of monocyte-stimulated IL-6, TNF-α, CCL4, and higher perceived stress compared to retired widow(er)s. We also found that family income was inversely associated with perceived stress and grief for employed widow(er)s, such that employed widow(er)s with lower or middle incomes had higher perceived stress and higher grief symptoms. Among retired widow(er)s, there was no relationship between income and psychological health (i.e., perceived stress and grief symptoms). For high-income bereaved employees, work may be pleasurable and rewarding; therefore, they may choose to stay employed for the status and extra discretionary income. Employees in the lower and middle levels of income fared the worst with psychological health; they likely must work to support their families. Beyond some income-based differences, on average, employed widow(er)s had higher proinflammatory marker production and perceived stress than those who were retired. These inflammatory differences exist despite employed widowers being about ten years younger than their retired counterparts.

Our assumption that greater LPS-stimulated cytokine production is associated with poor mental and physical health outcomes was based on (a) observational studies showing that greater LPS-stimulated proinflammatory cytokine production is associated with stress, depression, fatigue, and pain (e.g., Davis et al., 2008; Prather et al., 2007, 2009), (b) randomized clinical trial showing that mind-body interventions that reduced LPS-stimulated cytokine production along with somatic symptoms such as fatigue (e.g., Kiecolt-Glaser et al., 2014). However, a recent study showed that mindfulness-based intervention led to more pronounced ex vivo cytokine production, suggesting greater production is adaptive (Lindsay et al., 2022). We suspect this discrepancy is due to third-variable influences that moderate the association between psychological stress and LPS-stimulated cytokine production (Lindsay et al., 2022). When exposed to a pathogen, monocytes must produce cytokines in order for the innate immune system to mount a suitable response. Nevertheless, an overactive or sustained proinflammatory response can produce a maladaptive chronic proinflammatory state by promoting glucocorticoid resistance and dysregulating other processes involved in a well-regulated immune system (Cohen et al., 2012; Miller et al., 2002). When analyzing the capacity of immune cells to produce inflammatory mediators after ex vivo stimulation, it is not possible to identify the degree of production representative of a dysregulated immune response relative to a well-regulated immune response. Our scientific premise that, under most conditions, greater LPS-stimulated cytokine production is associated with a dysregulated response is based on the extant literature showing significant positive linear relationships between LPS-stimulated cytokine and poor health outcomes (e.g., depression, fatigue; Davis et al., 2008; Kiecolt-Glaser et al., 2014; Prather et al., 2007, 2009). Nevertheless, given a certain degree of cytokine production is necessary for a healthy immune response, characteristics of the study cohort and contextual influences may lead to circumstances where study participants exhibiting less LPS-stimulated cytokine production have a more adaptive response relative to those with more cytokine production. Accordingly, caution is warranted when interpreting the health impact ex vivo cytokine production, especially based on a single study.

Employed widow(er)s had higher proinflammatory marker production and higher perceived stress in our study. Three pathways may explain these findings: an increase in objective work demands, a subjective evaluation of higher work demands, and fewer opportunities for work recovery after bereavement. In contrast to employed widow(er)s, widowed retirees do not need to accommodate increases in work demands or challenges around work recovery. Employees may have greater emotional labor after bereavement. Emotional labor, the process of masking genuine emotions according to the norms and regulations of a job, is generally considered harmful to employees' health (Grandey & Gabriel, 2015). Widow(er)s report a heavier burden of emotional regulation at work (e.g., avoiding crying at their workplace; Gilbert et al., 2021). Emotional labor post-bereavement increases widowed employees' work demands.

Employees may subjectively assess higher work demands after bereavement. In qualitative interviews, bereaved employees cite feeling unmotivated and distracted at work (Gilbert et al., 2021). The inability to focus on work may leave widowed employees feeling less competent. Lower feelings of competence were associated with higher feelings of grief, depression, and loneliness in a prior bereavement study (Utz et al., 2011).

Recently bereaved employees may need to meet higher family demands (e.g., planning a funeral) outside of their work hours. Higher family demands post-bereavement may affect employees' ability to rest after work and increase their fatigue (Bennett et al., 2018). Context-specific demands at work and home drain one's personal resources. Lower personal resources reduce one's ability to address future demands (ten Brummelhuis & Bakker, 2012). Lower personal resources also increasing one's likelihood of experiencing negative health outcomes (Gallo & Matthews, 2003). These three pathways (i.e., greater objective work demands, greater subjective work demands, and lower work recovery) may be related to group differences in proinflammatory marker production and perceived stress for employed widow(er)s compared to retired widow(er)s.

Family income was inversely related to grief symptoms and perceived stress for bereaved employees but not bereaved retirees. Unequal access to bereavement leave may perpetuate class-based psychological health disparities. Our findings align with prior research that indicates that health disparities decrease linearly as income rises (Adler et al., 1994). We found that employed people with high family income levels had comparable grief symptoms and perceived stress levels to their retired counterparts. Employees in the lower and middle levels of family income may be unable to afford unpaid bereavement leave; in contrast, high-income bereaved employees may be more able to take time off from work. Therefore, widow(er)s at the lower and middle levels of family income may not reduce their work demands.

In efforts to better support bereaved employees, organizations should offer bereavement leave, as researchers have demonstrated that employees use bereavement leave when it is accessible (McGuinness, 2009; Wilson et al., 2021). Alternatively, employers may consider offering paid family leave. Providing paid family leave that is not limited to employees’ parental functions (e.g., the birth or adoption of children) may be important for bereaved employees who cannot comfortably miss paychecks. Supervisors should know their organization's leave policies and support their employees’ use of leave (Gilbert et al., 2021). Supervisor support may increase employees' use of existing leave policies, among other organizational support resources (Dimoff & Kelloway, 2019). However, taking leave is not an all-encompassing solution for bereavement. Supervisors and other organizational representatives should (1) communicate with bereaved employees to inform them about resources available and ascertain their needs, (2) accommodate employee needs in an individualized return-to-work plan, (3) recognize the employee's loss and demonstrate a willingness to approach the subject, and (4) offer emotional support to the bereaved employee (i.e., the C.A.R.E. model, Gilbert et al., 2021).

Work demands may be perceived through the lens of an employee's social context, including their family life. Being employed significantly impacted proinflammatory marker production and perceived stress three months after spousal bereavement, despite the potentially positive effects of psychological resource gain from work, which may serve as a net benefit later in widowhood (Fitzpatrick & Bossé, 2001). More research is needed to understand when work is beneficial or harmful to widow(er)s. For high income workers, resources derived from work, such as routine, status, and social support, may be important to an employee coping with the destabilizing effects of bereavement. High income may afford them conveniences that offset the stress associated with the additional employment-related stress, which emphasizes the need for flexibility in how employees can use bereavement leave (Gilbert et al., 2021).

Health disparities exist each step down the socioeconomic status ladder (Bindley et al., 2019; Cohen et al., 2008). The reserve capacity model provides a conceptual connection between socioeconomic status (SES), psychosocial resources, stress, and health (Gallo & Matthews, 2003). The reserve capacity model builds on the conservation of resources theory in the context of social determinates to health. The models indicates that people are motivated to gather and maintain resources (e.g., skills, knowledge, social support, time); therefore, stress occurs when resources are threatened, lost, or not gained after investment (Hobfoll, 2001). According to the reserve capacity perspective, people in the lower and middle ends of the SES ladder may have fewer practical and tangible resources and therefore have smaller reserve capacities to address demands and stressors, which lead to downstream health outcomes such as cardiovascular morbidity and mortality (Gallo & Matthews, 2003). Demands at work and home may compound to affect one's reserve capacity. As with our findings, this implies that only employed widow(er)s with the highest income levels can accommodate the added demands of work without suffering higher inflammation or psychological stress. We found support for this perspective in our study, as only employees with high levels of income had comparable levels of perceived stress and grief with retired widow(er)s.

4.1. Limitations and Future Directions

This study has several strengths that help to inform the literature on bereavement, work, and health. We assessed widow(er)s' health within the first three months after bereavement and consistently collected blood samples in the morning. We also measure the capacity of immune cells to produce inflammatory markers after ex-vivo stimulation, which allow for a representation of the immune system’s in-vivo response to stress and infection (Zuiden et al., 2011). Finally, we consider the intersection of two contextual factors (i.e., employment status and family income) on health post-bereavement.

The present findings could serve as a catalyst for more find-grained work understanding the characteristics of employment in immunological outcomes. In the current study we examined the association between widow(er)s’ employment status in relation to mental and physical health outcomes. We did not have data available elucidating employees’ workplace characteristics, a vital next step for future investigations. Next, our analyses captured experiences cross-sectionally three months post-bereavement, which may not represent experiences at later time points. Employment positively affects male widowers' physical health at 1-3 years after bereavement (Fitzpatrick & Bossé, 2001). When considered with present findings, one might hypothesize that employed widow(er)s might have a different bereavement trajectory than retired widow(er)s. Perhaps workers follow experience higher immune dysregulation and stress immediately after bereavement but return to pre-bereavement levels quicker than retirees (e.g., the “recovered” pathway; Bonanno et al., 2011). Research should examine the longitudinal effects of working during widowhood in future work. Finally, our sample is racially homogenous (mostly white), attenuating the findings' external validity.

5. Conclusion

This paper contributes to the literature in three ways. To our knowledge, only two prior studies have investigated the role of employment status among widow(er)s’ health (Koeneman et al., 2012; Pai & Barrett, 2007). No prior research has examined inflammation and employment status among widowers. Our analyses demonstrate the significant post-loss effects of employment on widow(er)s’ health. Next, we examined the association between employment status and the interaction between employment status and socioeconomic status. Our findings imply employed widow(er)s have different post-bereavement experiences depending on their socioeconomic status. Widow(er) s' socioeconomic status determines whether they experience employment as a choice or a necessity. In the occupational health psychology literature, there has been scant work on the effects of bereavement despite calls for more research (Charles-Edwards, 2009; Gilbert et al., 2021; Hazen, 2009; Wilson et al., 2021). In the psychoneuroendocrinology literature, studies have predominately focused on individual differences related to life events, and social relationships, with a considerable focus on people's context outside of work (Irwin, 2008). It is critical for future research to account for the impact that employment status and workplace context impact autonomic, neuroendocrine, and immune processes. Our goal is for the present study to serve as a catalyst for more fine-grained future work in this area.

Three months after bereavement employed widow(er)s had higher proinflammatory marker production and perceived stress than their retired counterparts. Employment may be a feature underlying risk and resilience for some widow(er)s. Our results suggest value in examining the tandem associations of work and family with inflammatory and psychological health.

Supplementary Material

MMC1

Highlights.

  • Widowed workers had higher proinflammatory marker production than widowed retirees

  • Widowed workers had higher perceived stress than widowed retirees

  • Widowed workers with lower and middle levels of income may be more vulnerable

  • Employment may be a feature underlying risk and resilience for some widow(er)s

Author Note:

This project was supported by funding from the National Institutes of Health (R01HL127260, PI Christopher Fagundes). JP (1F32AG079624-01), MC (5F31AG069439-02), LWC (1F31AG074648-01), AL (K01AG073824), and CP (1R01AG062690, 1R01AG062690-02S1, 1R21AG061597-01A1) are funded by the National Institute on Aging. Please direct communication about this article to Jensine Paoletti (Jensine.Paoletti@Rice.edu).

Conflicts of Interest

The authors have no conflicts of interest to disclose. This project was supported by funding from the National Institutes of Health (R01HL127260, PI Christopher Fagundes).

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

References

  1. Adler NE, Boyce T, Chesney MA, Cohen S, Folkman S, Kahn RL, & Syme SL (1994). Socioeconomic status and health: The challenge of the gradient. American Psychologist, 49(1), 15–24. 10.1037/0003-066X.49.1.15 [DOI] [PubMed] [Google Scholar]
  2. Adler NE, Epel ES, Castellazzo G, & Ickovics JR (2000). Relationship of subjective and objective social status with psychological and physiological functioning: Preliminary data in healthy, White women. Health Psychology, 19(6), 586–592. 10.1037/0278-6133.19.6.586 [DOI] [PubMed] [Google Scholar]
  3. Akiyama H, Barger S, Barnum S, Bradt B, Bauer J, Cole GM, Cooper NR, Eikelenboom P, Emmerling M, Fiebich BL, Finch CE, Frautschy S, Griffin WST, Hampel H, Hull M, Landreth G, Lue L, Mrak R, Mackenzie IR, … Wyss-Coray T (2000). Inflammation and Alzheimer’s disease. Neurobiology of Aging, 21(3), 383–421. 10.1016/S0197-4580(00)00124-X [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Bakker AB, ten Brummelhuis LL, Prins JT, & der Heijden F. M. M. A. van. (2011). Applying the job demands–resources model to the work–home interface: A study among medical residents and their partners. Journal of Vocational Behavior, 79(1), 170–180. 10.1016/j.jvb.2010.12.004 [DOI] [Google Scholar]
  5. Bindley K, Lewis J, Travaglia J, & DiGiacomo M (2019). Disadvantaged and disenfranchised in bereavement: A scoping review of social and structural inequity following expected death. Social Science & Medicine, 242, 112599. 10.1016/j.socscimed.2019.112599 [DOI] [PubMed] [Google Scholar]
  6. Bonanno GA, Westphal M, & Mancini AD (2011). Resilience to loss and potential trauma. Annual Review of Clinical Psychology, 7(1), 511–535. 10.1146/annurev-clinpsy-032210-104526 [DOI] [PubMed] [Google Scholar]
  7. Bruenig D, Morris CP, Young RM, & Voisey J (2014). Stress and resilience in combat-related PTSD: Integration of psychological theory and biological mechanisms. In Martin CR, Preedy VR, & Patel VB (Eds.), Comprehensive Guide to Post-Traumatic Stress Disorder (pp. 1–19). Springer International Publishing. 10.1007/978-3-319-08613-2_110-1 [DOI] [Google Scholar]
  8. Buysse DJ, Reynolds CF, Monk TH, Berman SR, & Kupfer DJ (1989). The Pittsburgh Sleep Quality Index: A new instrument for psychiatric practice and research. Psychiatry Research, 28(2), 193–213. 10.1016/0165-1781(89)90047-4 [DOI] [PubMed] [Google Scholar]
  9. Carballo-Penela A, Varela J, & Bande B (2019). The direct and indirect effects of self-efficacy on salespeople's emotional exhaustion and work-family conflict: A study using the Job Demands-Resources Model. Canadian Journal of Administrative Sciences / Revue Canadienne Des Sciences de l’Administration, 36(3), 363–376. 10.1002/cjas.1503 [DOI] [Google Scholar]
  10. Charles-Edwards D (2009). Empowering people at work in the face of death and bereavement. Death Studies, 33(5), 420–436. 10.1080/07481180902805632 [DOI] [PubMed] [Google Scholar]
  11. Charlson ME, Pompei P, Ales KL, & MacKenzie CR (1987). A new method of classifying prognostic comorbidity in longitudinal studies: Development and validation. Journal of Chronic Diseases, 40(5), 373–383. 10.1016/0021-9681(87)90171-8 [DOI] [PubMed] [Google Scholar]
  12. Chirinos DA, Ong JC, Garcini LM, Alvarado D, & Fagundes C (2019). Bereavement, self-reported sleep disturbances and inflammation: Results from Project HEART. Psychosomatic Medicine, 81(1), 67–73. 10.1097/PSY.0000000000000645 [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Cohen S, Alper CM, Doyle WJ, Adler N, Treanor JJ, & Turner RB (2008). Objective and subjective socioeconomic status and susceptibility to the common cold. Health Psychology, 27(2), 268–274. 10.1037/0278-6133.27.2.268 [DOI] [PubMed] [Google Scholar]
  14. Cohen S, Janicki-Deverts D, Doyle WJ, Miller GE, Frank E, Rabin BS, & Turner RB (2012). Chronic stress, glucocorticoid receptor resistance, inflammation, and disease risk. Proceedings of the National Academy of Sciences, 109(16), 5995–5999. 10.1073/pnas.1118355109 [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Cohen S, Kamarck T, & Mermelstein R (1983). A global measure of perceived stress. Journal of Health and Social Behavior, 24(4), 385–396. 10.2307/2136404 [DOI] [PubMed] [Google Scholar]
  16. Davis MC, Zautra AJ, Younger J, Motivala SJ, Attrep J, & Irwin MR (2008). Chronic stress and regulation of cellular markers of inflammation in rheumatoid arthritis: Implications for fatigue. Brain, Behavior, and Immunity, 22(1), 24–32. 10.1016/j.bbi.2007.06.013 [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. De Carlo A, Girardi D, Falco A, Dal Corso L, & Di Sipio A (2019). When does work interfere with teachers' private life? An application of the Job Demands-Resources Model. Frontiers in Psychology, 10, 1121. 10.3389/fpsyg.2019.01121 [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. DiGiacomo M, Lewis J, Phillips J, Nolan M, & Davidson PM (2015). The business of death: A qualitative study of financial concerns of widowed older women. BMC Women's Health, 15(1), 36. 10.1186/s12905-015-0194-1 [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Dimoff JK, & Kelloway EK (2019). With a little help from my boss: The impact of workplace mental health training on leader behaviors and employee resource utilization. Journal of Occupational Health Psychology, 24(1), 4–19. 10.1037/ocp0000126 [DOI] [PubMed] [Google Scholar]
  20. Eisma MC, Stroebe MS, Schut HAW, van den Bout J, Boelen PA, & Stroebe W (2014). Development and psychometric evaluation of the Utrecht Grief Rumination Scale. Journal of Psychopathology and Behavioral Assessment, 36(1), 165–176. 10.1007/s10862-013-9377-y [DOI] [Google Scholar]
  21. Erdogan B, Bauer TN, Truxillo DM, & Mansfield LR (2012). Whistle while you work: A review of the life satisfaction literature. Journal of Management, 38(4), 1038–1083. 10.1177/0149206311429379 [DOI] [Google Scholar]
  22. Fagundes CP, & Way B (2014). Early-life stress and adult inflammation. Current Directions in Psychological Science, 23(4), 277–283. 10.1177/0963721414535603 [DOI] [Google Scholar]
  23. Fagundes CP, & Wu EL (2020). Matters of the heart: Grief, morbidity, and mortality. Current Directions in Psychological Science, 29(3), 235–241. 10.1177/0963721420917698 [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Fagundes CP, Brown RL, Chen MA, Murdock KW, Saucedo L, LeRoy A, Wu EL, Garcini LM, Shahane AD, Baameur F, & Heijnen C (2019). Grief, depressive symptoms, and inflammation in the spousally bereaved. Psychoneuroendocrinology, 100, 190–197. 10.1016/j.psyneuen.2018.10.006 [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Fagundes CP, Murdock KW, LeRoy A, Baameur F, Thayer JF, & Heijnen C (2018). Spousal bereavement is associated with more pronounced ex vivo cytokine production and lower heart rate variability: Mechanisms underlying cardiovascular risk? Psychoneuroendocrinology, 93, 65–71. 10.1016/j.psyneuen.2018.04.010 [DOI] [PubMed] [Google Scholar]
  26. Fitzpatrick TR, & Bossé R (2001). Employment and health among older bereaved men in the normative aging study. Social Work in Health Care, 32(2), 41–60. 10.1300/J010v32n02_03 [DOI] [PubMed] [Google Scholar]
  27. Funk JL, & Rogge RD (2007). Testing the ruler with item response theory: Increasing precision of measurement for relationship satisfaction with the Couples Satisfaction Index. Journal of Family Psychology, 21(4), 572–583. 10.1037/0893-3200.21.4.572 [DOI] [PubMed] [Google Scholar]
  28. Gallo LC, & Matthews KA (2003). Understanding the association between socioeconomic status and physical health: Do negative emotions play a role? Psychological Bulletin, 129(1), 10–51. 10.1037/0033-2909.129.1.10 [DOI] [PubMed] [Google Scholar]
  29. Gallo LC, Bogart LM, Vranceanu A-M, & Matthews KA (2005). Socioeconomic status, resources, psychological experiences, and emotional responses: A test of the reserve capacity model. Journal of Personality and Social Psychology, 88(2), 386–399. [DOI] [PubMed] [Google Scholar]
  30. Gilbert S, Mullen J, Kelloway EK, Dimoff J, Teed M, & McPhee T (2021). The C.A.R.E. model of employee bereavement support. Journal of Occupational Health Psychology, 26(5), 405–420. 10.1037/ocp0000287 [DOI] [PubMed] [Google Scholar]
  31. Goh J, Pfeffer J, & Zenios SA (2016). The Relationship Between Workplace Stressors and Mortality and Health Costs in the United States. Management Science, 62(2), 608–628. 10.1287/mnsc.2014.2115 [DOI] [Google Scholar]
  32. Gouin J-P, Glaser R, Loving TJ, Malarkey WB, Stowell J, Houts C, & Kiecolt-Glaser JK (2009). Attachment avoidance predicts inflammatory responses to marital conflict. Brain, Behavior, and Immunity, 23(7), 898–904. 10.1016/j.bbi.2008.09.016 [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Grandey AA, & Gabriel AS (2015). Emotional labor at a crossroads: Where do we go from here? Annual Review of Organizational Psychology and Organizational Behavior, 2(1), 323–349. [Google Scholar]
  34. Hazen MA (2009). Recognizing and responding to workplace grief. Organizational Dynamics, 38(4), 290–296. 10.1016/j.orgdyn.2009.07.002 [DOI] [Google Scholar]
  35. Holmes TH, & Rahe RH (1967). The social readjustment rating scale. Journal of Psychosomatic Research, 11(12), 213–218. [DOI] [PubMed] [Google Scholar]
  36. Irwin MR (2008). Human psychoneuroimmunology: 20 Years of discovery. Brain, Behavior, and Immunity, 22(2), 129–139. 10.1016/j.bbi.2007.07.013 [DOI] [PubMed] [Google Scholar]
  37. Juth V, Smyth JM, Carey MP, & Lepore SJ (2015). Social constraints are associated with negative psychological and physical adjustment in bereavement. Applied Psychology: Health and Well-Being, 7(2), 129–148. 10.1111/aphw.12041 [DOI] [PubMed] [Google Scholar]
  38. Kamerāde D, Wang S, Burchell B, Balderson SU, & Coutts A (2019). A shorter working week for everyone: How much paid work is needed for mental health and well-being? Social Science & Medicine, 241, 112353. 10.1016/j.socscimed.2019.06.006 [DOI] [PubMed] [Google Scholar]
  39. Kenis G, & Maes M (2002). Effects of antidepressants on the production of cytokines. International Journal of Neuropsychopharmacology, 5(4), 401–412. [DOI] [PubMed] [Google Scholar]
  40. Kiecolt-Glaser JK, Bennett JM, Andridge R, Peng J, Shapiro CL, Malarkey WB, Emery CF, Layman R, Mrozek EE, & Glaser R (2014). Yoga’s impact on inflammation, mood, and fatigue in breast cancer survivors: A randomized controlled trial. Journal of Clinical Oncology, 32(10), 1040–1049. 10.1200/JCO.2013.51.8860 [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Koeneman MA, Chinapaw MJ, Verheijden MW, van Tilburg TG, Visser M, Deeg DJ, & Hopman-Rock M (2012). Do major life events influence physical activity among older adults: The Longitudinal Aging Study Amsterdam. International Journal of Behavioral Nutrition and Physical Activity, 9(1), 147. 10.1186/1479-5868-9-147 [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Knowles LM, Ruiz JM, & O’Connor M-F (2019). A Systematic Review of the Association Between Bereavement and Biomarkers of Immune Function. Psychosomatic Medicine, 81(5), 415–433. 10.1097/PSY.0000000000000693 [DOI] [PubMed] [Google Scholar]
  43. LeRoy AS, Gabert T, Garcini L, Murdock KW, Heijnen C, & Fagundes CP (2020). Attachment orientations and loss adjustment among bereaved spouses. Psychoneuroendocrinology, 112. 10.1016/j.psyneuen.2019.104401 [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Levine CS, Hoffer LC, & Chen E (2017). Moderators of the relationship between frequent family demands and inflammation among adolescents. Health Psychology, 36(5), 493–501. 10.1037/hea0000469 [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Lin T, Sung Y, Syu J, Lin K, Hsu H, Liao M, Liu Y, & Lin S (2020). Anti-inflammatory and antiarrhythmic effects of beta blocker in a rat model of rheumatoid arthritis. Journal of the American Heart Association, 9(18), e016084. [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Lindsay EK, Creswell JD, Stern HJ, Greco CM, Walko TD, Dutcher JM, Wright AGC, Brown KW, & Marsland AL (2022). Mindfulness-based stress reduction increases stimulated IL-6 production among lonely older adults: A randomized controlled trial. Brain, Behavior, and Immunity, 104, 6–15. 10.1016/j.bbi.2022.05.001 [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Mancini AD, Robinaugh D, Shear K, & Bonanno GA (2009). Does attachment avoidance help people cope with loss? The moderating effects of relationship quality. Journal of Clinical Psychology, 65(10), 1127–1136. 10.1002/jclp.20601 [DOI] [PubMed] [Google Scholar]
  48. McGuinness B (2009). Grief in the workplace. Bereavement Care, 28(1), 2–8. 10.1080/02682620902746037 [DOI] [Google Scholar]
  49. Miller GE, Cohen S, & Ritchey AK (2002). Chronic psychological stress and the regulation of pro-inflammatory cytokines: A glucocorticoid-resistance model. Health Psychology, 21, 531–541. 10.1037/0278-6133.21.6.531 [DOI] [PubMed] [Google Scholar]
  50. Moon JR, Glymour MM, Vable AM, Liu SY, & Subramanian SV (2014). Short- and long-term associations between widowhood and mortality in the United States: Longitudinal analyses. Journal of Public Health, 36(3), 382–389. 10.1093/pubmed/fdt101 [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. Nersesian PV, Han H-R, Yenokyan G, Blumenthal RS, Nolan MT, Hladek MD, & Szanton SL (2018). Loneliness in middle age and biomarkers of systemic inflammation: Findings from Midlife in the United States. Social Science & Medicine, 209, 174–181. 10.1016/j.socscimed.2018.04.007 [DOI] [PMC free article] [PubMed] [Google Scholar]
  52. O’Connor MF (2019). Grief: A Brief History of Research on How Body, Mind, and Brain Adapt. Psychosomatic Medicine. 10.1097/PSY.0000000000000717 [DOI] [PMC free article] [PubMed] [Google Scholar]
  53. O'Connor M-F, Bower JE, Cho HJ, Creswell JD, Dimitrov S, Hamby ME, Hoyt MA, Martin JL, Robles TF, Sloan EK, Thomas KS, & Irwin MR (2009). To assess, to control, to exclude: Effects of biobehavioral factors on circulating inflammatory markers. Brain, Behavior, and Immunity, 23(7), 887–897. 10.1016/j.bbi.2009.04.005 [DOI] [PMC free article] [PubMed] [Google Scholar]
  54. Pai M, & Barrett AE (2007). Long-term payoffs of work? Women’s past involvement in paid work and mental health in widowhood. Research on Aging, 29(5), 436–456. 10.1177/0164027507304084 [DOI] [Google Scholar]
  55. Pawlowski T, Radkowski M, Małyszczak K, Inglot M, Zalewska M, Jablonska J, & Laskus T (2014). Depression and neuroticism in patients with chronic hepatitis C: Correlation with peripheral blood mononuclear cells activation. Journal of Clinical Virology, 60(2), 105–111. 10.1016/j.jcv.2014.03.004 [DOI] [PubMed] [Google Scholar]
  56. Prather AA, Marsland AL, Muldoon MF, & Manuck SB (2007). Positive affective style covaries with stimulated IL-6 and IL-10 production in a middle-aged community sample. Brain, Behavior, and Immunity, 21(8), 1033–1037. 10.1016/j.bbi.2007.04.009 [DOI] [PMC free article] [PubMed] [Google Scholar]
  57. Prather AA, Marsland AL, Hall M, Neumann SA, Muldoon MF, & Manuck SB (2009). Normative variation in self-reported sleep quality and sleep debt is associated with stimulated pro-inflammatory cytokine production. Biological Psychology, 82(1), 12–17. 10.1016/j.biopsycho.2009.04.008 [DOI] [PMC free article] [PubMed] [Google Scholar]
  58. Rohleder N (2014). Stimulation of systemic low-grade inflammation by psychosocial stress. Psychosomatic Medicine, 76(3), 181–189. [DOI] [PubMed] [Google Scholar]
  59. Roulston A, Campbell A, Cairnduff V, Fitzpatrick D, Donnelly C, & Gavin A (2017). Bereavement outcomes: A quantitative survey identifying risk factors in informal carers bereaved through cancer. Palliative Medicine, 31(2), 162–170. 10.1177/0269216316649127 [DOI] [PubMed] [Google Scholar]
  60. Schultze-Florey CR, Martínez-Maza O, Magpantay L, Breen EC, Irwin MR, Gündel H, & O’Connor M-F (2012). When grief makes you sick: Bereavement induced systemic inflammation is a question of genotype. Brain, Behavior, and Immunity, 26(7), 1066–1071. 10.1016/j.bbi.2012.06.009 [DOI] [PMC free article] [PubMed] [Google Scholar]
  61. Stahl ST, Arnold AM, Chen J-Y, Anderson S, & Schulz R (2016). Mortality after bereavement: The role of cardiovascular disease and depression. Psychosomatic Medicine, 78(6), 697–703. 10.1097/PSY.0000000000000317 [DOI] [PMC free article] [PubMed] [Google Scholar]
  62. Stroebe MS, Schut H. a. W., & Stroebe W (2007). Health consequences of bereavement: A review. The Lancet Infectious Diseases, 370(9603), 1960–1973. 10.1016/S0140-6736(07)61816-9 [DOI] [PubMed] [Google Scholar]
  63. Stroebe M, & Schut H (1999). The dual process model of coping with bereavement: Rationale and description. Death Studies, 23(3), 197–224. 10.1080/074811899201046 [DOI] [PubMed] [Google Scholar]
  64. Toossi M, & Torpey E (2017). Older workers: Labor force trends and career options (Career Outlook). U.S. Bureau of Labor Statistics. https://www.bls.gov/careeroutlook/2017/article/older-workers.htm [Google Scholar]
  65. Utz RL, Lund DA, Caserta MS, & deVries B (2011). Perceived self-competency among the recently bereaved. Journal of Social Work in End-of-Life & Palliative Care, 7(2–3), 173–194. 10.1080/15524256.2011.593154 [DOI] [PMC free article] [PubMed] [Google Scholar]
  66. Willerson JT, & Ridker PM (2004). Inflammation as a Cardiovascular Risk Factor. Circulation, 109(21_suppl_1), II–2. 10.1161/01.CIR.0000129535.04194.38 [DOI] [PubMed] [Google Scholar]
  67. Wilson DM, Punjani S, Song Q, & Low G (2021). A study to understand the impact of bereavement grief on the workplace. OMEGA - Journal of Death and Dying, 83(2), 187–197. 10.1177/0030222819846419 [DOI] [PubMed] [Google Scholar]
  68. Zoccola PM, & Dickerson SS (2012). Assessing the relationship between rumination and cortisol: A review. Journal of Psychosomatic Research, 73(1), 1–9. 10.1016/j.jpsychores.2012.03.007 [DOI] [PubMed] [Google Scholar]
  69. van Zuiden M, Heijnen CJ, Schoot R. van de, Amarouchi K, Maas M, Vermetten E, Geuze E, & Kavelaars A (2011). Cytokine production by leukocytes of military personnel with depressive symptoms after deployment to a combat-zone: A prospective, longitudinal study. PLOS ONE, 6(12), e29142. 10.1371/journal.pone.0029142 [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

MMC1
NIHMS1867510-supplement-MMC1.docx (71.6KB, docx)

RESOURCES